Project Abstract Background: Attrition of participants is a critical concern for investigators conducting HIV prevention intervention studies with people who inject drugs (PWID) and their intimate partners. Eastern Europe and Central Asia, the only region globally where HIV incidence and mortality is still increasing among PWID, is a crucial setting for HIV prevention and treatment. Kazakhstan has one of the largest numbers of PWID in the world with an incidence rate of HIV that has grown precipitously in recent years. The HIV epidemic is concentrated among PWID and their sexual partners in Kazakhstan. The growth of HIV among PWID through drug risk behaviors has fueled transmission of HIV to sexual partners and has resulted in rapid expansion of the HIV epidemic. Rigorous HIV prevention intervention studies for PWID that incorporate couples in Kazakhstan are crucial to attenuating the HIV epidemic in Central Asia. Research suggests that criminal justice involvement (CJI)?i.e., arrest, detention and incarceration?is a barrier to PWID being retained in HIV interventions and accessing HIV and drug services and treatment. PWID in Kazakhstan have especially high rates of CJI, impacting approximately 90% of PWID. No studies to date have examined the impact of CJI of PWID and their intimate partners on attrition in behavioral HIV prevention intervention studies. To address these gaps, this application has two research aims: 1) To examine relationships between CJI and attrition, among PWID and their intimate partners and 2) To examine dyadic relationships between partners' CJI and attrition during the study period. Methods: This dissertation study uses a sequential explanatory mixed methods approach with 1) secondary data from Project Renaissance, a study of 300 male PWID plus their female partners who were randomized to a 5-session Risk Reduction (RR) or Wellness Promotion (WP) intervention arm and re-assessed at 3, 6 and 12 months and 2) primary qualitative follow-up data with PWID and their intimate partners to inform findings from the secondary data analysis. Analysis: The Actor-Partner Interdependence Model (APIM) using Structural Equation Modeling (SEM) will estimate the effects of CJI on attrition after adjusting for potential confounders (AIM 1 and 2). The analytic framework provided by APIM is an innovative method of identifying patterns within dyads of how PWID and their intimate partners' CJI shape attrition in relation to each other within a single model (AIM 2). Primary data collection and qualitative analysis will enrich statistical findings from the APIM using SEM. The 2-year research and training program proposed in this NRSA (F31) Fellowship application will cultivate skills in mixed-methods research that are essential to completing my dissertation research and training and will provide critical academic professional skills for a career as an independently funded substance abuse and HIV prevention researcher. !